Clinical Nutrition 34 (2015) 335e340

Contents lists available at ScienceDirect

Clinical Nutrition
journal homepage: http://www.elsevier.com/locate/clnu

ESPEN endorsed recommendation

Diagnostic criteria for malnutrition e An ESPEN Consensus Statement

T. Cederholm a, *, I. Bosaeus b, R. Barazzoni c, J. Bauer d, A. Van Gossum e, S. Klek f,
M. Muscaritoli g, I. Nyulasi h, J. Ockenga i, S.M. Schneider j, M.A.E. de van der Schueren k, l,
P. Singer m
a

Departments of Geriatric Medicine, Uppsala University Hospital and Public Health and Caring Sciences, Clinical Nutrition and Metabolism, Uppsala
University, Uppsala, Sweden
b
Clinical Nutrition Unit, Sahlgrenska University Hospital and University of Gothenburg, Gothenburg, Sweden
c
Department of Medical, Surgical and Health Sciences, University of Trieste, Trieste, Italy
d
t, Oldenburg, Germany
Department of Geriatric Medicine, Carl von Ossietzky Universita
e
Department of Gastroenterology, Clinic of Intestinal Diseases and Nutritional Support, Hopital Erasme, Free University of Brussels, Brussels, Belgium
f
General and Oncology Surgery Unit, Stanley Dudrick's Memorial Hospital, Skawina, Poland
g
Department of Clinical Medicine, Sapienza University of Rome, Rome, Italy
h
Department of Nutrition and Dietetics and Department of Medicine, Monash University Central Clinical School, Prahran, Australia
i
Department of Gastroenterology, Hepatology, Endocrinology, and Nutrition, Klinikum Bremen Mitte, Bremen, Germany
j
Department of Gastroenterology and Clinical Nutrition, University Hospital and University of Nice Sophia-Antipolis, Nice, France
k
Department of Nutrition and Dietetics, Internal Medicine, VU University Medical Center, Amsterdam, The Netherlands
l
Department of Nutrition, Sports and Health, Faculty of Health and Social Studies, HAN University of Applied Sciences, Nijmegen, The Netherlands
m
Department of General Intensive Care, Institute for Nutrition Research, Rabin Medical Center, Sackler School of Medicine, Tel Aviv University, Petah Tikva
49100, Israel

a r t i c l e i n f o

s u m m a r y

Article history:
Received 27 February 2015
Accepted 3 March 2015

Objective: To provide a consensus-based minimum set of criteria for the diagnosis of malnutrition to be
applied independent of clinical setting and aetiology, and to unify international terminology.
Method: The European Society of Clinical Nutrition and Metabolism (ESPEN) appointed a group of
clinical scientists to perform a modied Delphi process, encompassing e-mail communications, face-toface meetings, in group questionnaires and ballots, as well as a ballot for the ESPEN membership.
Result: First, ESPEN recommends that subjects at risk of malnutrition are identied by validated
screening tools, and should be assessed and treated accordingly. Risk of malnutrition should have its own
ICD Code. Second, a unanimous consensus was reached to advocate two options for the diagnosis of
malnutrition. Option one requires body mass index (BMI, kg/m2) <18.5 to dene malnutrition. Option
two requires the combined nding of unintentional weight loss (mandatory) and at least one of either
reduced BMI or a low fat free mass index (FFMI). Weight loss could be either >10% of habitual weight
indenite of time, or >5% over 3 months. Reduced BMI is <20 or <22 kg/m2 in subjects younger and older
than 70 years, respectively. Low FFMI is <15 and <17 kg/m2 in females and males, respectively. About 12%
of ESPEN members participated in a ballot; >75% agreed; i.e. indicated
7 on a 10-graded scale of
acceptance, to this denition.
Conclusion: In individuals identied by screening as at risk of malnutrition, the diagnosis of malnutrition
should be based on either a low BMI (<18.5 kg/m2), or on the combined nding of weight loss together
with either reduced BMI (age-specic) or a low FFMI using sex-specic cut-offs.
2015 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

Keywords:
Malnutrition
Nutritional assessment
Body composition
Denition
Consensus
Delphi

1. Introduction

* Corresponding author. Tel.: 46 702733192.

E-mail address: tommy.cederholm@pubcare.uu.se (T. Cederholm).

Malnutrition due to starvation, disease or ageing can be dened

as a state resulting from lack of uptake or intake of nutrition
leading to altered body composition (decreased fat free mass) and
body cell mass leading to diminished physical and mental function

http://dx.doi.org/10.1016/j.clnu.2015.03.001
0261-5614/ 2015 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.

336

T. Cederholm et al. / Clinical Nutrition 34 (2015) 335e340

and impaired clinical outcome from disease [1]. Although this

denition is well-accepted, the condition lacks clear and generally
accepted diagnostic criteria.
During the famine catastrophes in Africa during the 1960s, WHO
brought attention to the medical aspects of starvation [2]. The
concepts of kwashiorkor and marasmus were introduced to dene
a particular protein decient condition characterized by hypoalbuminemic peripheral edema and ascites, i.e. kwashiorkor, and
a particular energy decient state characterized by severe weight
loss due to mainly fat store depletion, i.e. marasmus. This classication did not turn out as relevant for the recognition and diagnosis
of malnutrition that was increasingly observed in hospitals in the
Western countries during the later decades of the last century.
Since depletion is usually a combined deciency and loss of protein
and energy, the general term of protein-energy malnutrition (PEM)
became widely accepted [3]. Clinical characteristics used to dene
PEM have varied over time, but there was never a consensus on
diagnostic criteria. Various combinations of clinical, anthropometrical, biochemical and immunological measures were used (e.g.
Refs. [4,5]).
The last decades have seen the advent of several malnutrition
screening tools that have reached increased acceptance due to their
clinical feasibility. These screening tools combine about the same
variables, i.e. weight loss, body mass index (BMI), signs of eating
difculties (e.g. appetite loss or reduced food intake) and a grading
of on-going disease severity. ESPEN recognizes the following risk
screening tools to be used in the hospital, elderly care and community settings; Nutritional Risk Screening 2002 (NRS-2002), Mini
Nutritional Assessment-Short Form (MNA-SF) and Malnutrition
Universal Screening Tool (MUST) [6]. The major use of these tools is
to screen for malnutrition risk. The subsequent clinical actions
should be assessment of underlying mechanisms and type of
nutritional problems, in order to design personalized nutritional
therapies. The diagnostic procedure usually ends there. The reasons
for this are many-fold. One may be the imperative to focus on the
implementation of the malnutrition risk screening procedures over
the past years. Another apparent reason is the lack of consensus
over diagnostic criteria. In the absence of such criteria, it is difcult
to distinguish the effectiveness and efcacy of nutritional therapies
when applied in different phases of the patients decline into
malnutrition. The effects of nutritional therapy given at an early
stage, before body protein and energy stores have been depleted,
might differ as compared to when given at a late stage with overt
depletion.
Previous important consensus initiatives have been attempted
[7e9], and the outcomes have reected the complexity of the issue
and the difculties to reaching consensus. It was considered that
these results may fuel the discussion within the nutritional
societies, which will most ideally lead to an international consensus
on a denition and operationalism of malnutrition [7].
There is moreover a confusion of terminology. Malnutrition,
protein-energy malnutrition, undernutrition, depletion, wasting,
cachexia are some of the terms used to denominate the condition
that ensues deciencies of macro- and micronutrients and catabolism of protein and energy stores due to disease and ageing. The
malnutrition-related concepts cachexia [10,11], sarcopenia [11,12]
and frailty [13] are today well-established. The current initiative
doesn't challenge their denitions.
This statement is aimed at helping clinicians to effectively provide therapeutic nutritional interventions, and accordingly, to
document clinically relevant malnutrition, and moreover to facilitate documentation in the Disease Related Group (DRG) and International Classication of Diseases (ICD) systems.
Malnutrition should be recognized as a serious clinical risk
factor. In real clinical settings this is not the case, partly due to the

lack of simple and unequivocal diagnostic criteria. It is crucial to

reach consensus on diagnostic criteria for malnutrition in order to
unify the terminology (e.g. for ICD-10), to enhance the legitimacy of
nutritional practices, to improve clinical care and to move the
clinical and scientic nutrition eld forward. For this purpose the
European Society of Clinical Nutrition and Metabolism (ESPEN)
appointed a consensus group with a clear mission to provide
criteria for the diagnosis of malnutrition.

1.1. The consensus group objectives

The primary objective was to reach consensus for simple, clear
and generally applicable diagnostic criteria of malnutrition in the
sense of energy and protein store depletion. The intention was to
provide criteria that are independent from etiologic mechanisms,
and that can be used for all patients and in all clinical settings.
Secondary objectives were to try to bring clarity to the nutritional terminology, and to provide a general nutritional disorders
concept tree.

2. Methods
2.1. Design of the delphi process and selection of the expert group
ESPEN decided in late 2012 to launch the initiative. In January
2013 representatives of more than 40 member countries of ESPEN
outlined the initiative and acknowledged the process. An international expert group consisting of experienced clinical scientists was
gathered to perform a modied Delphi process. The consensus
group participants, i.e. the authors behind this report, were chosen
to represent the clinical elds of medicine, surgery, intensive care,
oncology and geriatrics.
It was agreed within the group to base the process on open email communications, face-to-face meetings and on open as well as
closed ballots within the group. The intention was to maintain the
communication at each step until consensus for each milestone
(see below) was reached among all participants before the next
step was taken. Furthermore, the group agreed to seek the opinions
of the ESPEN members before deciding on the statement. Finally, it
was decided to perform validation studies of the nal statement.
This paper reports the process and outcome of the Delphi process,
i.e. the consensus based malnutrition diagnostic criteria, while the
validation studies are on-going and will be presented separately.

2.2. Dened milestones of the process

Overall there were ve major milestones dened for the project.
These were to
- decide the interrelationship between the screening and the
diagnostic process.
- identify the individual criteria that cover the crucial components of the condition for all patients in all settings. For the sake
of feasibility, the goal was to use as few criteria as possible.
- decide on whether and how the individual criteria should be
combined to provide an expected high specicity of the
diagnosis.
- come up with clinically relevant cut-off values based on relevant
reference populations for the chosen individual criteria.
Finally, an independent milestone was to decide whether the
term of undernutrition or malnutrition is to be preferred.

T. Cederholm et al. / Clinical Nutrition 34 (2015) 335e340

3. Results
3.1. How are screening and the diagnostic procedures interrelated?
There was a strong consensus from e-mail communications, a
one-day face-to-face meeting, and an anonymous group ballot that
the diagnostic procedure has to be based on the outcome of a
screening evaluation. Nutritional screening is always mandatory in
all clinical and care settings, since it is unanimously recognized that
patients affected by acute and chronic diseases are at high risk of
developing nutritional impairment. Diagnostic measures are only
needed for those cases that score positive for nutritional risk by any
of the validated screening tools. This consensus group does not
recommend any specic of the validated tools, as long as the tool is
validated for the setting where it is applied. Screening should be
sensitive, whereas diagnosis is specic. Thus, fewer will be diagnosed as suffering from malnutrition than the number of subjects
that are identied as being at risk. It was acknowledged that
measurements that were performed during the screening process
could potentially be used also for the diagnostic decision.
Risk of malnutrition was suggested to be as a diagnosis with
its own ICD Code. The general acceptance of the prevailing
malnutrition screening tools relies on the fact that fullling the
criteria for risk of malnutrition imposes negative clinical outcomes,
including death. Therefore it is crucial to commence nutritional
therapy as early as possible. Such intervention generates costs.
Therefore, the diagnosis of being at risk of malnutrition needs to
be coded and reimbursed in the ICD and DRG systems.
3.2. Which individual criteria do best capture the state of
malnutrition?
Nutritional variables are traditionally categorized as measures of
anthropometry, including body composition, biochemical indicators and data related to eating capability. The latter include
history as well as recording of actual food intake, or information on
factors for potential limitations of food intake like anorexia,
dysphagia or chewing problems.
A one-day face-to-face meeting was organized to scrutinize
these potential diagnostic criteria. Functional measures of strength
or power were soon dismissed as part of the diagnosis since they
were consensually viewed as not nutrition specic enough. Thus,
the following variables, or group of variables, were discussed as
potential criteria.

337

3.2.5. Reduced food intake

Self-reported or quantied reduction of the food intake is a
feature of most of the screening tools.
3.2.6. Biochemical indicators
Markers of visceral protein status, like serum concentrations of
albumin have a tradition of being used as markers of nutritional
status. Inammation, due to disease or ageing, is likely the most
common cause underlying the development of malnutrition. Thus,
inammation is an important etiologic factor for malnutrition.
3.2.7. Subjective professional evaluation
After the screening procedure has identied subjects at nutritional risk, a subjective evaluation of body composition and physical function performed by experienced health professionals could
indicate the diagnosis of malnutrition.
3.3. The result of the criteria ballot in the consensus group
As a result of the consensus group face-to face meeting, a
questionnaire was constructed and provided to all group members
where they were asked to grade their degree of acceptance from 1
to 6 to include the following variables as one of the diagnostic
criteria; i.e. weight loss, reduced BMI, reduced fat free mass index
(FFMI), reduced fat mass index (FMI), reduced food intake, reduced
appetite, a biochemical marker (visceral proteins or inammatory
markers), or a subjective evaluation of a professional. The response
of the questionnaire was anonymous for all but the facilitator (TC).
The scores of 1e2, 3e4 and 5e6 were denoted against, middle and
pro, respectively. Figure 1 shows the result of the ballot. A fairly
clear pattern emerged, as there was a preference in the group for
the use of weight loss, reduced BMI and reduced FFMI.
The ensuing e-mail discussion conrmed that the term weight
loss would encompass signicant reductions in food intake, appetite loss, and for any imbalance between nutrient requirements and
intake. Thus, evaluations of food intake and appetite will not add
further information beyond weight loss. Nevertheless, data on
intake and appetite are important for screening for malnutrition
risk.
Next BMI was discussed. Low BMI is commonly accepted in the
medical community, and received support from the majority of the
consensus group to be one criterion. However, BMI can be
misleading as the growing obesity epidemic renders a number of
seriously ill and malnourished patients in the normal BMI range
although they may have lost a considerable and clinically relevant

3.2.1. Anthropometry
BMI and leg and arm anthropometry, e.g. calf circumference,
arm muscle circumference and triceps skinfold, are clinically
available objective variables with a fairly wide-spread use.
3.2.2. Body composition
Fat free mass (FFM) and fat mass (FM) can objectively be
measured by technical devices like bioelectrical bio-impedance
analyses (BIA), dual energy x-ray absorptiometry (DXA),
computed tomography (CT), ultrasound or magnetic resonance
imaging (MRI).
3.2.3. Weight loss
This reects a dynamic process that requires a negative energy
balance; i.e. a reduced food intake or increased energy expenditure.
3.2.4. Anorexia
Loss of appetite is a common complication to disease, medication and ageing. It is one of the most important mechanisms behind
weight loss.

Fig. 1. Outcome of the questionnaire on the preference of the individual criteria. Group
members were asked to grade from 1 to 6; 1 total disagreement and 6 total
agreement, on the use of each criterion. Scores of 1e2 was aggregated to being against,
scores of 5e6 as being for (pro) and 3e4 as in between (middle).

338

T. Cederholm et al. / Clinical Nutrition 34 (2015) 335e340

amount of body mass. Due to several question marks on the use of

BMI for malnutrition diagnosis some of the consensus group
members suggested to omit BMI as a potential criterion. Nevertheless in the nal analysis it was unanimously decided to keep BMI
for the nal selection, due to the fact that BMI has a well-validated
outcome predictive value.
For body composition, the pros and cons of fat mass and fat free
mass to reect nutrition status was discussed. One incentive for
ESPEN to launch this malnutrition diagnosis initiative was to test
the grounds for promoting body composition measures for diagnostic purposes. Information on relative amounts of fat mass and
fat free mass provides far more valuable information than BMI
alone, both from functional and metabolic points of view. FFM, i.e.
mainly the muscle mass was considered to provide the most relevant information. Muscle confers ability to activity, acts as the
amino acid pool and is important for energy expenditure and
glucose metabolism [14].
According to biochemical indicators, the perception was that
visceral proteins like serum albumin concentrations are good indicators of disease severity and outcome. Inammation is today
considered the major reason for reduced serum levels of visceral
proteins. Thus, visceral proteins should not be used for either
screening or diagnosis of malnutrition. The low grade of nutrition
specicity was the major reason for this decision. The question of
integrating a biochemical indicator of inammatory activity, e.g. Creactive protein, into the diagnostic format raised another discussion. Inammation has a central etiologic role for catabolism and is
an integral feature of cachexia. Thus, inammation is rather to be
regarded as an etiologic factor than a diagnostic feature of
malnutrition. The same reasoning goes for indicators of potential
anabolic activity, e.g. insulin-like growth factor-1.
The Consensus group decided that all criteria should be objective as far as possible. Diagnosis by subjective evaluation by an
experienced health professional did not meet this criterion.
3.4. Is there an optimal combination of criteria?
Thus, three variables were chosen to most accurately reect
malnutrition, namely weight loss, reduced BMI and reduced FFMI.
The next question was whether and how to combine the variables
to best capture the diagnosis.
For this purpose the group performed a second anonymous
ballot based on a questionnaire that comprised nine options. For
each option, any of the three favored variables was suggested to be
mandatory and combined with one or both of the other two variables. Eight out of ten answers were positive for the option where
unintentional weight loss should be mandatory, and weight loss
should be combined with either reduced BMI or with reduced FFMI.
Consensus was reached to suggest both alternatives, i.e. either
weight loss and reduced BMI, or weight loss and reduced FFMI,
should give the diagnosis of malnutrition. In real clinical life access
to body composition measurements is limited, which justies
keeping both alternatives.
3.5. Which are the clinically relevant cut-off values for the chosen
individual criteria?
The next challenge was to nd valid and clinically relevant cutoff values for weight loss, BMI and FFMI.
3.5.1. Weight loss
Weight loss trajectories differ with clinical condition. Nevertheless, involuntary weight loss is a strong predictor of negative
outcomes irrespective of magnitude, speed and underlying cause.
Naturally, a massive and fast weight loss due to an aggressive

cancer disease imposes a higher risk than a smaller and slower

weight loss due to ageing.
Thus, consensus was reached to propose two optional cut-offs
for unintentional weight loss; i.e. either >5% over the last 3
months to cover for acute illnesses, or >10% of habitual weight
indenite of time to be relevant for chronic conditions.

3.5.2. BMI
WHO advocates BMI <18.5 kg/m2 as a general cut-off for underweight. This cut-off is justied at a public health population
level [15], whereas its relevance for clinical and care settings may
be questioned. As already mentioned the trend of increasing BMI in
all populations world-wide make this acknowledged BMI cut-off
value difcult to use for the purpose of dening malnutrition. Patients struck with highly catabolic diseases may in 3e6 months lose
substantially more than 10% of their weight and still have BMI
values well above normal ranges. Another issue to consider is that
epidemiological evidence indicates that older populations display
higher optimal BMI intervals (e.g. for survival) than younger people.
Partly due to the strong global acceptance of the WHO cut-off of
18.5 kg/m2 it was decided unanimously to accept the WHO recommended cut-off of as a criterion that in its own right will be
enough to diagnose malnutrition.
With this latter decision it was easy to come to consensus for a
complementary suggestion for relevant BMI cut-off values; namely
<20 kg/m2 for subjects <70 years of age, and <22 kg/m2 for subjects
70 years and older, remembering the fact that these BMI levels need
to be linked to weight loss as dened above. The choices of 20 and
22 kg/m2, respectively, were based on consensus in the group.
Ethnic and regional variability in BMI may need to be considered.

3.5.3. FFMI
Cut-offs for FFMI need to be linked to the decided cut-offs for
BMI on one hand, and to the fact that women have lower FFMI (and
higher FMI) than men on the other hand. Based on Swiss reference
material [16] it was decided to suggest FFMI <15 and <17 kg/m2 in
women and men, respectively. It has to be emphasized that reference values, like for BMI, should be relevant for the specic ethnic
and cultural context that is at hand.
The result of this process is summarized in the Fact box.
Fact box: Two alternative ways to diagnose malnutrition.
Before diagnosis of malnutrition is considered it is
mandatory to fulfil criteria for being at risk of malnutrition
by any validated risk screening tool.
Alternative 1:
 BMI <18.5 kg/m2
Alternative 2:
 Weight loss (unintentional) > 10% indefinite of time, or
>5% over the last 3 months combined with either
 BMI <20 kg/m2 if <70 years of age, or <22 kg/m2 if
70
years of age or
 FFMI <15 and 17 kg/m2 in women and men, respectively.
3.6. ESPEN membership ballot
For transparency, implementation and awareness purposes the
nal consensus group suggestion for the malnutrition diagnostic

T. Cederholm et al. / Clinical Nutrition 34 (2015) 335e340

339

poll among the consensus group. Malnutrition had a slight preponderance with 53% of the votes as compared to 47% in favor of
undernutrition. Due to this uncertain result the consensus group
doesn't advocate any specic term, but has chosen to use malnutrition for this paper.
Finally, a conceptual tree of prevailing nutritional disorders was
constructed (Fig. 3), acknowledging the complex interactions between the conditions. The structure of Fig. 3 was discussed as some
in the consensus group preferred a less hierarchical arrangement of
the conditions.
4. Discussion

Fig. 2. ESPEN membership ballot. Vote count on the question We want you to indicate
on a scale from 1 to 10 whether you agree strongly 10, or disagree strongly 1, with
this suggestion (see Fact box).

criteria was presented to the membership of ESPEN to vote for. The

ESPEN Newsletter provided the questionnaire by e-mail to all
members (~around 3000 world-wide). Members were asked to
grade their acceptance of the suggestion, as presented in the Fact
box, on a scale from 1 (very low) to 10 (very high). Altogether 344
members gave their vote. Figure 2 shows the result of the poll; i.e.
>75% of the voting members gave a score of 7 or more in agreement
of the suggestion. A score of 4 or less indicating disagreement was
given by 10% of the voters. According to the given comments from
the members that gave low scores, the absence of functional
measures was the major reason for disagreement.
3.7. Terminology of malnutrition or undernutrition and the
conceptual tree of nutritional disorders
The general perception of the group was that malnutrition and
undernutrition is about equally used in the scientic literature and
in clinical practice, with a slight preponderance for malnutrition. A
potential problem with the term malnutrition is that it literally
covers all deviating nutritional states. In the rst in-group ballot
performed, the group members were asked to anonymously state
their preference. The two terms received an equal number of votes,
i.e. 5 vs 6. The pro and con arguments were not strongly in favor of
any of the terms and it was decided to leave the question open for
the ESPEN member ballot to give a stronger advice.
Thus, in the same ESPEN Newsletter that encouraged ESPEN
members to vote for or against the suggested diagnostic criteria
(see above), the members were also asked to give their opinion on
whether they preferred the term of malnutrition or of undernutrition. The vote came out as indecisive and split as in the smaller

Fig. 3. A conceptual tree of nutritional disorders.

This ESPEN Consensus Statement provides a novel, comprehensive, yet simple, format for the diagnosis of malnutrition. The
statement acknowledges the well-accepted BMI cut-off of 18.5 kg/
m2 provided by WHO. In addition, it introduces an alternative and
partly new format based on the combination of unintentional
weight loss, and low BMI or low FFMI, where the latter drives the
necessity to use modern techniques for body composition measurements. Regarding BMI, there was unanimous consensus that, in
the presence of signicant weight loss, BMI levels higher than
18.5 kg/m2 may also reect sufciently altered nutritional state to
warrant the diagnosis of malnutrition. The proposed cutoff values
of 20 (<70 y) and 22 (
70 y) kg/m2 was the result from consensus
within the group. Validation studies have been launched to possibly
conrm these indications.
Recent years have seen the development of several malnutrition
screening tools, for example NRS-2002, MNA-SF and MUST, which
should be used for everybody that comes in contact with health and
elderly care. The purpose is to identify the individuals that will
need further nutritional assessment and that will need nutritional
therapy, in order to prevent the further nutritional deterioration
into malnutrition. To be dened as at risk for malnutrition according to any of the validated screening tools should render its
own ICD Code. Many of the validated screening tools in clinical use
today also provide grading into manifest malnutrition as well,
which is somewhat beyond their primary purpose. Since NRS-2002,
MNA-SF and MUST use different criteria and cut-offs, and were
designed for different purposes and populations [17], the prevalence of malnutrition in a specic population will differ with the
screening tool used. Moreover, as indicated the appropriate evaluation of nutritional therapy effects is affected by when the therapy
is applied, i.e. in late or early phases of malnutrition, as well as by
variations in the diagnostic criteria [18]. These examples illustrate
how acknowledgment and development of clinical nutrition may
suffer from the lack of clear diagnostic criteria for malnutrition.
As indicated the purpose of this initiative was to provide a
general diagnosis that is relevant for all subjects in all clinical settings. Recent years have seen the advent of improved conceptualization and better understanding of different forms of nutritional
disorders and malnutrition (Fig. 3). Nevertheless, the denition is
meant to be valid for all individuals no matter if starvation, acute or
chronic disease or ageing is the dominant underlying cause of the
state of malnutrition. As with all general diagnoses (compare
anemia) they need to be elaborated into sub-diagnoses which
indicate their etiologies to facilitate the best treatment. Thus, this
ESPEN statement also provides a conceptual tree where pure starvation, disease-related malnutrition (cachexia) [10,11], and the
concepts of sarcopenia [11,12] and frailty [13] are related to, and at
least partly over-arched by the general term of malnutrition.
The concepts of cachexia and sarcopenia pinpoint the importance of variations in body composition, i.e. fat and muscle mass, as
predictors of clinical outcome. The two-compartment model for
body composition, i.e. FMI and FFMI, offers a more precise

340

T. Cederholm et al. / Clinical Nutrition 34 (2015) 335e340

description of the investigated subject than for example BMI, and is

thus providing a clearer justication for clinical decisions. Therefore, it was encouraging to see the strong consensus of the group to
advocate the potential requirement of FFMI for the diagnosis of
malnutrition. Devices for body composition measurements, e.g.
bioelectrical impedance analyzers and DXA machines, are not
readily available for daily routine care. Thus it was crucial to not to
mandate FFMI for the diagnosis of malnutrition. This Consensus
Statement will hopefully drive and promote the need to develop
accessible techniques for body composition measurements in all
health care settings, i.e. at hospital wards, in primary care, in
nursing homes as well as in home elderly care.
Beyond the identication of diagnostic criteria this initiative
also aimed to clarify whether the term malnutrition or undernutrition is to be preferred. As reported above the vote, in the
consensus group as well as among the voting ESPEN membership,
was more or less split in two halves. For the moment there is a small
preponderance for the use of malnutrition in the nutrition community. The uncertain outcome of this issue doesn't justify any
strong recommendation.
In conclusion, this ESPEN Consensus Statement has rendered a
clear and simple format for the diagnosis of malnutrition. Parallel
initiatives are on-going. The strong objective of the current ESPEN
statement is to provide malnutrition diagnostic criteria that are
independent of etiology, whereas other initiatives state that they
aim to provide etiology-driven criteria [8,9]. The various global
initiatives need to be linked and in the long run it is crucial to reach
a global consensus on this important issue. On-going and future
validation studies will provide data that may conrm the feasibility
of, or be used to update, the malnutrition criteria of this ESPEN
Consensus Statement.
Conict of interest
None of the authors display any CoI in the production of this
manuscript. Travel expenses for the one-day meeting were reimbursed by ESPEN. No other grants from any funding agency were
offered.
References
[1] Sobotka L, editor. Basics in clinical nutrition. 4th ed. Galen; 2012.
[2] Jeliffe DB. The assessment of the nutritional status of the community.. Geneva:
World Health Organization; 1966. Monograph Series No. 53.

[3] Waterlow JC, Tomkins AM, Grantham-McGregor SM. Protein energy malnutrition. London: Edward Arnold; 1992.
[4] Buzby GP, Mullen JL, Matthews DC, Hobbs CL, Rosato EF. Prognostic nutritional
index in gastrointestinal surgery. Am J Surg 1980;139:160e7.
[5] Detsky AS, McLaughlin JR, Baker JP, Johnston N, Whittaker S, Mendelson RA,
et al. What is subjective global assessment of nutritional status? JPEN J
Parenter Enter Nutr 1987;11:8e13.
[6] Kondrup J, Allison SP, Elia M, Vellas B, Plauth M. Educational and clinical
practice committee, European society of parenteral and enteral nutrition
(ESPEN). ESPEN guidelines for nutrition screening 2002. Clin Nutr 2003;22:
415e21.
[7] Meijers JM, van Bokhorst-de van der Schueren MA, Schols JM, Soeters PB,
Halfens RJ. Dening malnutrition: mission or mission impossible? Nutrition
2010;26:432e40.
[8] Jensen GL, Mirtallo J, Compher C, Dhaliwal R, Forbes A, Grijalba RF, et al. International consensus guideline committee. Adult starvation and diseaserelated malnutrition: a proposal for etiology-based diagnosis in the clinical
practice setting from the International consensus guideline Committee. Clin
Nutr 2010;29:151e3.
[9] White JV, Guenter P, Jensen G, Malone A, Schoeld M. Academy malnutrition
work Group; A.S.P.E.N. Malnutrition task force; A.S.P.E.N. Board of Directors.
Consensus statement: academy of nutrition and dietetics and american society for parenteral and enteral Nutrition: characteristics recommended for the
identication and documentation of adult malnutrition (undernutrition).
J Parenter Enter Nutr 2012;36:275e83.
s J, Bales C, Baracos V, Guttridge D, et al. Cachexia:
[10] Evans WJ, Morley JE, Argile
a new denition. Clin Nutr 2008;27:793e9.
s J, Aversa Z, Bauer JM, Biolo G, et al. Consensus
[11] Muscaritoli M, Anker SD, Argile
denition of sarcopenia, cachexia and pre-cachexia: joint document elaborated by Special Interest Groups (SIG) cachexia-anorexia in chronic wasting
diseases and nutrition in geriatrics. Clin Nutr 2010;29:154e9.
[12] Cruz-Jentoft AJ, Baeyens JP, Bauer JM, Boirie Y, Cederholm T, Landi F, et al.
European Working Group on Sarcopenia in older people. Sarcopenia: European consensus on denition and diagnosis: report of the European Working
Group on Sarcopenia in Older People. Age Ageing 2010;39:412e23.
[13] Fried LP, Tangen CM, Walston J, Newman AB, Hirsch C, Gottdiener J, et al.
Cardiovascular health study collaborative research group. Frailty in older
adults: evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;56:
M146e56.
[14] Biolo G, Cederholm T, Muscaritoli M. Muscle contractile and metabolic
dysfunction is a common feature of sarcopenia of aging and chronic diseases:
from sarcopenic obesity to cachexia. Clin Nutr 2014;33:737e48.
[15] WHO. Physical status: the use and interpretation of anthropometry. Report of
a WHO expert committee. World Health Organization technical report series
854. 1995.
[16] Schutz Y, Kyle UU, Pichard C. Fat-free mass index and fat mass index percentiles in Caucasians aged 18-98 y. Int J Obes Relat Metab Disord 2002;26:
953e60.
[17] van Bokhorst-de van der Schueren MA, Guaitoli PR, Jansma EP, de Vet HC.
Nutrition screening tools: does one size t all? A systematic review of
screening tools for the hospital setting. Clin Nutr 2014;33:39e58.
[18] Milne AC, Potter J, Vivanti A, Avenell A. Protein and energy supplementation
in elderly people at risk from malnutrition. Cochrane Database Syst Rev
2009;15(2):CD003288.